Returns from Investing in Cryptocurrency: Evidence from German Individual Investors - Blockchain ...
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
BRL Working Paper Series No. 6
Returns from Investing in Cryptocurrency: Evidence from
German Individual Investors
Lennart Ante 1, 2, *, Ingo Fiedler 1, 2, 3, Marc von Meduna 1, 2, Fred Steinmetz 1, 2
1 Blockchain Research Lab, Colonnaden 72, 22303 Hamburg
2 University of Hamburg, Faculty of Business, Economics & Social Sciences, Von-Melle-
Park 5, 20146 Hamburg, Germany
3
Concordia University, Faculty of Arts & Science, 2070 Mackay Street, Montreal, QC,
H3G 2J1, Canada
* Correspondence: ante@blockchainresearchlab.org
Published: 19 Feb 2020
Abstract: Cryptocurrencies such as Bitcoin are a high-risk asset class where
very high returns are offset by large losses. In a nationally representative online
survey of 3,864 German citizens, 354 (9.2%) reported owning cryptocurrencies
in March 2019. We analyze a subpopulation of 225 cryptocurrency owners who
classify as investors. 56% of them experienced positive returns, while 29% had
negative results. The remaining respondents broke even. The average investment
was €1,773 in a portfolio of two cryptocurrencies. At the time of the survey, the
average portfolio value had risen to €7,094 – an average gain of 300%. While
nearly half of the investors (44%) outperformed Bitcoin market returns, not a
single one of the early investors (2009-12) did. We find that net income, the
degree of cryptocurrency knowledge and the degree of ideological motivation
for owning cryptocurrency have positive effects on returns. This first scientific
analysis of individual investment in cryptocurrencies provides a basis for future
research and for regulatory decision-making.
Keywords: Bitcoin; Blockchain; Individual investors; Alternative investments;
Financial performance
1 Introduction
Following the publication of the Bitcoin whitepaper and the subsequent launch of the Bitcoin
network, the price of the digital currency rose from zero to almost $20,000 in December 2017
and then dropped back to $4,000. Thus, investing in Bitcoin is clearly risky. While Nakamoto
(2008) introduced Bitcoin as electronic cash, that simple characterization no longer applies, as
the benefits and applications have expanded and shifted significantly over the last ten years.
Additionally, thousands of other cryptocurrencies with different focuses and features have
emerged since, which are characterized by even higher returns and higher risk of loss alike.
However, besides media reports about e.g. Bitcoin millionaires or total losses due to scams,comparatively little is known about the extent to which investments in cryptocurrencies have
actually benefited or harmed retail investors.
Cryptocurrencies are used as digital cash, as a medium of exchange and as a store of value.
Furthermore, so-called tokens, specific types of cryptocurrency, can represent additional
characteristics, like vouchers, licenses, voting rights or securities (Ante and Fiedler 2019). Most
cryptocurrency transactions are attributable to speculators, rather than to ‘users’ of the currency
(Yermack 2015). Evidence of price bubbles underlines the speculative nature of
cryptocurrencies (Corbet, Lucey, et al. 2018; Kristoufek 2018). Other research suggest that, due
to the comparatively low level of regulation (cryptocurrencies are not financial instruments),
price manipulation and informed trading are more likely to occur (Ante 2019; Feng et al. 2018).
Baur et al. (2018) find that Bitcoin has similarities to stocks, bonds and commodities, and
suggest that Bitcoin is a speculative investment. Dyhrberg (2016) analyzes the financial asset
capabilities of Bitcoin and in conclusion places Bitcoin somewhere between gold and the dollar.
Other research suggests that Bitcoin has no intrinsic value at all (Cheah and Fry 2015).
Exploring trading dynamics and market structures, Dyhrberg et al. (2018) show that Bitcoin is
investible for retail investors.
A range of research has dealt with the adoption of cryptocurrencies as a means of payment
(Schuh and Shy 2016), perceptions of cryptocurrency value (Gibbs and Yordchim 2014), the
determinants of interest in cryptocurrencies (F. Glaser et al. 2014), speculative opportunities
(Hur et al. 2015), the choice of specific cryptocurrencies (Al Shehhi et al. 2014) and the
relationship between cryptocurrency trading and gambling (Mills and Nower 2019). Few
studies have so far dealt with individual investors in cryptocurrency, but they have examined
issues related to investors into initial coin offerings (ICOs), blockchain-based crowd-financing
of startups (Ante et al. 2018). (Fisch et al. (2019) focus on the investment motives of 517 ICO
investors and find that technological motives are most important, followed by financial and
ideological motives. Fahlenbrach and Frattaroli (2019) analyze ICO investment behavior based
on Ethereum wallet data, finding that large investors buy their positions at a discount and sell
them profitably. However, none of these studies use population data on the socioeconomic
background of cryptocurrency investors and their actual returns. To date there is no research on
the performance of individual investments in cryptocurrency.
This paper aims to extend the literature on cryptocurrencies, their adoption and the question
as to whether cryptocurrencies are suitable investments by characterizing cryptocurrency
investors and the outcomes they achieve. We investigate whether investment success in
cryptocurrency is subject to similar explanatory factors as investment in similar asset classes.
Our data basis is derived from on a representative online survey among the German adult
population that comprises 3,864 individuals. 354 (9.2%) of the respondents reported owning
cryptocurrencies, but some of them 1) obtained their cryptocurrency not through investment
but e.g. through mining or 2) failed to provide sufficient information. This leaves a group of
225 (5.8%) individual cryptocurrency investors. We address questions regarding socio-
demographics, such as whether cryptocurrency investors differ from the rest of the population
with respect to their income or education. Additionally, we ask what returns the cryptocurrency
investors have achieved and whether they were able to beat the market. Do cryptocurrency
investors classify as short-term speculators or do they look towards the long term? The literature
on individual stock investors offers a suitable basis and benchmark. Our explorative study
2extends this literature by analyzing investors, classifying them according to their success and
using multivariate analysis to identify factors that influence investment success.
2 Literature and hypotheses
Barber and Odean (2013) point out that financial markets (and therefore also cryptocurrency
markets) are subject to an adding-up constraint: Whenever someone buys something, someone
else must sell it. Thus, the success of one investor necessitates the failure of another, which
makes trading a zero-sum game. If transaction costs, like trading fees, are taken into account,
most individual investors underperform the market. The returns achieved by individual stock
investors have been estimated at 1.5% below the market index, with underperformance of 6.5%
for active investors (Barber and Odean 2000).
H1: Individual investors underperform in comparison to market returns.
Yet any aggregate statistics on underperformance mask the huge differences among individual
investors. We therefore also investigate whether the investors’ socio-demographics affect their
success. The literature shows that men generally trade more often than women and achieve
lower average returns, while both underperform the market (Barber and Odean 2001). Anderson
(2013) finds that various socio-demographic characteristics of individual stock investors affect
their investment performance. Lower income, wealth, age and education are associated with
less diversified portfolios, higher trading frequency and lower financial performance. However,
Korniotis and Kumar (2011) find that investment performance tends to decline with age.
According to Grinblatt et al. (2012), stock bought by investors with a high IQ generates high
positive returns over investment horizons of up to one month. Similarly, well-educated
investors outperform others by up to 3.6% annually (Korniotis and Kumar 2013).
H2: Socio-demographics, which we operationalize as a) gender, b) age, c) income and d) the
level of education affect the financial performance from investing in cryptocurrency.
Barber and Odean (2013) ask why so many individual investors self-manage their portfolios
when they could use cheaper alternatives such as index funds, which moreover achieve higher
returns. Various potential explanations for the irrational behavior or underperformance of
individual investors have been proposed, including overconfidence and an illusion of control
(Barber and Odean 2002), biased self-assessment (M. Glaser and Weber 2007; Graham et al.
2009), sensation-seeking (Grinblatt and Keloharju 2009) or financial inexperience. Individual
investors generally have a home bias, i.e. they prefer assets that they are familiar with, ignoring
the fact that foreign stocks offer diversification advantages (French and Poterba 1991).
Investors who are associated with a certain industry tend to prefer securities from the same
industry, despite the resulting lack of diversification (Døskeland and Hvide 2011). A high self-
assessment of competence, as elicited for example through questions about investment products
or the assessment of opportunities on the market, is associated with higher trading frequency
and greater portfolio diversification (Graham et al. 2009).
H3: Industry knowledge, which we operationalize as a) the level of knowledge about
cryptocurrencies, b) the number of different cryptocurrencies known and c) the number of
different cryptocurrency exchanges used, affects the returns from investing in cryptocurrency.
3H4: Self-assessed attitudes in terms of a) the level of trust in cryptocurrencies and b) the
importance of ideological reasons to own cryptocurrencies affects the financial performance
from investing in cryptocurrency.
Diversification helps to reduce the risks regarding investment outcomes. However, individual
investors often do not diversify their portfolios, holding only a few different assets – four on
average, according to Barber and Odean (2000). Goetzmann and Kumar (2008) show that
typical portfolio volatility is unnecessarily high because the correlation among the assets held
exceeds the value that would result from randomly selected assets. Ivković et al. (2008) find
that a lower level of individual investors’ portfolio concentration is associated with higher
financial performance. Diversification affects the variability of performance, but not average
performance. We control for potential effects of diversification by testing if the number of
different cryptocurrencies owned or the number of different cryptocurrencies owned divided
by the sum of investment leads to relevant effects.
3 Sample and variables
The data used in this study is based on an online survey conducted from February 8 to March
28, 2019, that is representative of the German adult internet-using population in terms of age
and gender. Of the 3,846 respondents, 354 (9.2%) reported owning cryptocurrencies. 225
(5.8%) of them, having provided sufficient information on the size and time of their investment
and their current portfolio value, constitute the sample for this study. The survey participants
were recruited via mail, received no prior indication as to the topic of the survey, and were
financially incentivized to answer the questions. For descriptive analyses of the complete
sample we refer to (Blockchain Research Lab 2019a, 2019b).
The respondents were asked to indicate the value of their portfolio at the time of the survey and
the amount of their total investment in Euro, excluding any subsequent additional investments
based on cryptocurrency trading gains. Both numbers were divided by 1,000 to form the
variables portfolio and invest, respectively. The first dependent variable, return, is calculated
as: (portfolio – invest) / invest. As return is highly skewed and has positive, negative and zero
values, we calculate return (log) as: log (return +1) and use this variable instead for the
multivariate analysis. Additionally, we categorize the respondents broadly according to their
investment success using the dependent dummy variables positive, negative and no change.
Dummy variables are used to indicate the year of the first investment (2009 to 2019) –
respondents stated the month and year of their first investment. coins known refers to the
number of coins known to an investor, while coins owned represents the number of different
cryptocurrencies that an investor owns – the respondents were asked which ones of the 15
cryptocurrencies with the greatest market capitalization are currently in their portfolios.
Dummy variables for individual coin possession were also generated for descriptive analysis
(for the full list of the 15 cryptocurrencies, see Table 3). The variable coins / invest is calculated
as coins owned / invest. We use its log for the regression models (coins / invest (log)).
The respondents were furthermore asked to indicate to what extent they use their
cryptocurrency for each of two purposes: speculation (short-term) and investment (long-term).
Depending on which purpose is assigned greater importance, the respondents are classified as
1) short-term and 2) long-term investors. Where the two purposes received the same weight,
the respondents were classified as 3) indifferent investors.
4The participants were also asked to assess their knowledge of cryptocurrencies (crypto
knowledge) on a scale of 0 (no knowledge) to 10. exchanges indicates the number of different
cryptocurrency exchanges an investor uses. trust refers to the investors’ assessment, on a scale
of 0 to 10, of their trust in cryptocurrencies, while ideology captures, on the same scale, the
degree to which ideology motivates the investors to own cryptocurrencies.
Finally, we collected data on a number of socio-demographic variables: Age (in years);
gender (male); and monthly income, which was recorded in categories, each of which was in
turn assigned a value as follows: ‘under €500’ (0.45); ‘€500-999’ (0.75); ‘€1,000-1,499’ (1.25);
‘€1,500-1,999’ (1.75); ‘€2,000-2,999’ (2.5); ‘€3,000-4,999’ (4); ‘over €5,000’ (6.5). education
approximates a respondent’s level of education based on their highest educational attainment.
The variable assumes the following values: 0 for ‘no school leaving certificate’; 1 for ‘lower
secondary school’; 3 for ‘high school’; 3 for ‘commercial or trade training’; 4 for ‘university
degree’; 5 for ‘PhD’.
4 Results
4.1 Descriptive results
4.1.1 Investment characteristics and market returns
Table 1 summarizes investment characteristic for the 225 cryptocurrency investors according
to the year of their first investment. Each line presents the number of investors who entered the
market in a given year, their total investment and the sum of their portfolio values at the time
of the survey. Additionally, among each of these groups of investors, we distinguish three types
based on their investment horizon.
The returns achieved by the respondents between their first month and year of investment
and the end of March, 2019 (when the survey was conducted), are compared to the development
of the market return over the same period. Instead of a market index, as done in the stock
literature, we use buy-and-hold returns of Bitcoin. The reason for this is that there is no
continuous index over the entire period under consideration. Besides, Bitcoin is highly
correlated with the price of other cryptocurrencies (Katsiampa 2019). For the years 2009 to
2013, Bitcoin returns are only estimates, as we were unable to identify precise price histories.
From 2013 on, we relied on monthly close prices from coinmarketcap.com. The last column of
Table 1 shows the share of investors who underperformed compared to Bitcoin market returns
over the respective period.
The 225 investors invested a total of around €400,000. Today, their portfolios are worth
almost €1.6 million. Most investors entered the cryptocurrency market from 2015 onwards and
continued to do so at an increasing rate until 2018. If we extrapolate from the 8 investors who
first invested in the first three months of 2019, we obtain 24 new investors in 2019, which
suggests a declining interest in cryptocurrencies. The amounts invested have remained at a
comparatively high level since 2014. An extrapolation for all of 2019 would yield a sum of
€90,760, the highest value yet. Based on their self-assessment, 27% of the individuals are
classified as short-term investors, 41% have predominantly long-term investment interests, and
the remaining 32% are categorized as ‘indifferent’.
5Table 1. Cryptocurrency investment statistics by year of first investment
Investors Sum of Investor classification Investors returns
Sum of portfolio Share of under-
Market returns
Year investments values Short- Long- Weighted Absolute (Bitcoin) performers
# % (€) Indiff.
(€) term term average (€)
2009 2 1% 2,200 6,500 0% 50% 50% 195% 4,300 532,644,307% 100%
2010 5 2% 1,044 1,574 40% 20% 40% 51% 530 6,781,667% 100%
2011 11 5% 23,482 43,338 18% 45% 36% 85% 19,856 39,457% 100%
2012 5 2% 11,700 12,100 20% 20% 60% 3% 400 27,127% 100%
2013 11 5% 11,654 302,913 27% 64% 9% 2,499% 291,259 3,048% 91%
2014 11 5% 59,024 270,767 18% 27% 55% 359% 211,743 825% 73%
2015 22 10% 62,817 103,880 18% 50% 27% 65% 41,063 1,522% 91%
2016 31 14% 42,929 521,029 29% 42% 29% 1,114% 478,100 764% 68%
2017 58 26% 81,027 145,280 33% 41% 26% 79% 64,253 176% 50%
2018 61 27% 80,527 144,130 25% 38% 38% 79% 63,603 162% 20%
2019 8 4% 22,690 44,620 50% 38% 13% 97% 21,930 110% 63%
Σ 225 100% 399,094 1,596,131 27% 41% 32% 300% 1,197,037 4,888,368% 56%
Bitcoin prices from 2009 to 2012 are estimated annual averages (BTC/EUR ≈ 0.000000188 (2009), 0.000014746 (2010), 0.00286 (2011), 0.00369 (2012)); absolute
market index returns are based on the sum invested in a particular month and year.
6We find that very early investors, i.e. those who first invested in the years 2009 to 2012, have since
generated returns of between 195% (2011) and 51% (2013). The highest average relative returns
(2,499%) were generated by those who entered the market in 2013. The highest absolute return
accrues to those who first invested in 2016 (€478,100; an average of €15,422 per investor). Most of
the effect is attributable to two investors, one of whom invested €500 and realized €180,000 (age 38,
male, commercial/trade training, income €2,000-€2,999), while the other one invested €5,500 and
realized €185,000 (age 38, male, PhD, income above €5,000).
All those who first invested in the years 2009 to 2012 have theoretically underperformed the
market. We see the first ‘winners’ among those who entered in 2013, with 9% of them beating the
market. This share rose to 27% among those who first invested in 2014. The years of market entry
that produced the highest shares of ‘winners’ are 2017 (50%) and 2018 (80%). Clearly the market
index – as we define it – is hard to beat, considering that a Bitcoin investment of €100 in 2009 would
have yielded €531 million by the time of the survey. Of the 225 investors, 99 (44%) beat the market.
25 of them (25%) can be classified as short-term investors and 46 (46%) as long-term investors, the
rest being indifferent. With hindsight, those long-term investors clearly had the superior strategy,
considering the high buy-and-hold returns in the market.
4.1.2 Sample statistics and differences across investment outcomes
The average investment amounted to €1,772, which corresponds to 60.5% of the investors’ average
monthly disposable income of approximately €2,930. The average final portfolio value is €7,094 – a
return of 300%. The average portfolio contains just over two of the top 15 cryptocurrencies in terms
of market capitalization. Group III, i.e. those investors who achieved positive returns, has the lowest
average investment (€1,445) and the highest average portfolio value (€11,172), which corresponds to
a 673% increase in value.
The 225 investors are familiar with 5.92 of the top 15 cryptocurrencies on average. They rated
their knowledge about cryptocurrencies at an average of 7.58, with a comparatively small standard
deviation of 1.79 (cf. Table A2). The members of group III know the largest number of
cryptocurrencies (6.17), owning 2.09 of them, while group IV investors (negative returns) know only
6 cryptocurrencies. The difference in the number of known cryptocurrencies between the groups is
significant at the 5%-level. Group III investors consider themselves most knowledgeable about
cryptocurrencies (average self-assessment of 7.84), the difference to the less successful investors
(average of 7.25) again being significant.
Across the whole sample, the level of trust in cryptocurrencies was rated at an average of 6.79,
and a value of 6.16 was given for the ideological motivation to own cryptocurrencies. Regarding the
level of trust, groups II and III are very similar, while the value in group IV is markedly lower. The
difference between groups III and IV is significant at the 5%-level. A similar patter exists with respect
to the level of ideology: Groups II and III are close to each other but the difference between those
with positive and negative returns is significant at the 1%-level. It is conceivable that the investment
success of group III caused its members to place greater trust in cryptocurrency and to ascribe their
investment to ideology.
7Table 2. Sample statistics and differences across sub-samples
I. II. III. IV.
Δ
Full No Positive Negative III. vs. IV
sample change returns returns
Investment
Invest 1.77 1.78 1.45 2.59 1.15
Portfolio 7.09 1.78 11.17 1.50 9.67**
Industry knowledge
Crypto knowledge 7.58 7.26 7.84 7.25 0.60**
Coins known 5.92 4.85 6.17 6.00 0.17**
Exchanges 1.88 1.21 2.13 1.75 0.37
Attitudes
Trust 6.79 7.00 7.01 6.26 0.75**
Ideology 6.16 6.44 6.46 5.42 1.04***
Demographics
Age 37.64 38.82 37.21 37.83 0.62
Male 0.77 0.71 0.77 0.82 0.05
Income 2.93 2.74 3.13 2.64 0.48**
Education 3.01 2.82 3.06 3.02 0.05
Diversification
Coins owned 2.02 2.09 2.08 2.07 0.01
Coins / invest 0.03 0.03 0.06 0.03 0.03
Observations 225 34 126 65
% of full sample (n=3,846) 5.82% 0.87% 3.26% 1.69%
% of crypto investors (n=225) 100% 15% 56% 29%
***,**, * indicates significance at the 1%, 5% and 10% level, respectively (t-test).
Cryptocurrency investors are predominantly male (77%), 37.6 years old on average, have a monthly
income of €2,930 and an average education score of 3.01. The age distribution across the three
outcome groups is quite similar, with group II showing the highest average age at 38.82 and group
III the lowest at 37.21. The proportion of male investors is highest in the group with negative returns
and lowest in the group with positive returns, although the difference is insignificant. In terms of
income, however, the difference of €1,040 between groups III (€3,130) and II (€2,640) is significant
at the 5% level. Further statistics on subdivisions of the age, income and education metrics can be
found in Table A1. We identify is a highly significant difference of 16% in the €1,000 to €1,499 pay
grade between individuals with positive (9%) and negative (25%) returns.
4.1.3 Individual possession of different cryptocurrencies
While the previous sections have dealt only with cryptocurrencies in general, we now turn to
individual currencies. Table 3 shows how popular the 15 largest cryptocurrencies are within the
sample and the outcome groups. The selection of the top 15 cryptocurrencies was based on market
capitalization (on coinmarketcap.com). The respondents also had the opportunity to name any other
coins they owned beyond those 15. Ten other currencies were mentioned, of which only Dogecoin
(DOGE) was mentioned more than once.
8Table 3. Possession of the 15 ‘biggest’ cryptocurrencies among the sample and outcome groups
I. II. III. IV.
Δ
Full No Positive Negative III. vs. IV.
sample change returns returns
Bitcoin (BTC) 81.78% 70.59% 86.51% 78.46% 8.05%
Ethereum (ETH) 29.33% 23.53% 28.57% 33.85% 5.27%
Ripple (XRP) 20.00% 14.71% 20.63% 21.54% 0.90%
Bitcoin Cash (BCH) 13.78% 11.76% 16.67% 9.23% 7.44%
EOS (EOS) 3.11% 5.88% 3.17% 1.54% 1.64%
Stellar Lumens (XLM) 3.56% 5.88% 2.38% 4.62% 2.23%
Litecoin (LTC) 15.11% 11.76% 15.08% 16.92% 1.84%
Tether (USDT) 3.11% 2.94% 1.59% 6.15% 4.57%*
Bitcoin SV (BSV) 2.67% 5.88% 1.59% 3.08% 1.49%
TRON (TRX) 7.11% 5.88% 7.14% 7.69% 0.55%
Cardano (ADA) 1.78% 0.00% 1.59% 3.08% 1.49%
Iota (IOT) 7.11% 0.00% 8.73% 7.69% 1.04%
Monero (XMR) 6.22% 8.82% 5.56% 6.15% 0.60%
Binance Coin (BNB) 2.22% 0.00% 2.38% 3.08% 0.70%
Dash (DSH) 5.33% 0.00% 7.14% 4.62% 2.53%
Observations 225 34 126 65
* indicates significance at the 10% level (t-test).
The cryptocurrencies are ordered by their market capitalization. The largest currencies globally are
also the ones which are most widely held by the investors in our sample: Bitcoin (81.78%), Ethereum
(29.33%) and Ripple (20%). Litecoin (15.11%) is the fourth most popular currency in the sample and
ranks in seventh place based on market capitalization. Cardano (1.78%) and Binance Coin (2.22%)
are the least popular choices among the top 15 assets. EOS (3.11%) and Stellar Lumens (3.56%) are
likewise not very common in the sample despite being the fifth- and sixth-largest cryptocurrencies,
respectively. Iota, which ranks 12th in terms of global market capitalization, has its legal entity based
in Germany. This may explain its comparatively large prevalence in the sample (7.11%).
The popularity of the individual assets varies considerably across the three outcome groups, though
the difference between groups III and IV is significant only in the case of Tether (Δ = 4.57%; pthe prevalence is 7.44% percentage points higher in group III than in group IV. There is some connection with the distribution of Bitcoin across the groups as Bitcoin Cash is a hard fork: Every Bitcoin owner also received Bitcoin Cash in July 2017. On the other hand, the prevalence of Bitcoin SV, which is in turn a hard fork of Bitcoin Cash that was distributed in November 2018, is almost twice as high in group IV (3.08%) as in group III (Δ = 1.49). 5 Multivariate results Table A2 shows descriptive statistics, correlations between the variables and variance inflation factors (VIFs). Not surprisingly, the number of cryptocurrencies known correlates with the number of currencies owned (0.45; p
The effects we identify are largely consistent across the three outcome variables, though the level of significance varies. Models I and II show 10%-significant positive effects of crypto knowledge on returns. We also identify significant positive effects for income (p
to negative financial performance, and we caution that the self-assessed variables may be subject to
reverse causation: An unsuccessful investor may place less trust in cryptocurrencies or less
importance on ideology.
On average, investors hold about two cryptocurrencies in their portfolio – fewer than the average
of four stocks identified by Barber and Odean (2000). We fail to identify any significant effects of
portfolio concentration or diversification in terms of coins owned but find highly significant effects
of the ratio between coins owned and the sum of investments, which measures diversification as a
function of the amount invested. The results are in line with some of the existing literature (e.g.
Ivković et al. 2008). Our lack of effects in the respect of total coins owned may perhaps be explained
by the a high correlation between cryptocurrencies (Katsiampa 2017), which renders diversification
less relevant. This consideration may be supported by the ownership ratios of individual
cryptocurrencies (cf. Table 3).
The sum of investments of the cryptocurrency owners we surveyed average €1,773 or around 60%
of the respondents’ monthly income. Thus, most German retail investors do not appear to take
excessive financial and social risks when buying cryptocurrency, since even a total loss of the
investment would not threaten their existence. The comparatively low stakes may indicate a general
awareness that crypto assets entail serious risk. At least from the standpoint of consumer protection,
regulation of the industry thus does not seem imperative. This conclusion does not apply, however,
to crime that occurs within the cryptocurrency industry due to anonymity and irreversible transactions
(Ante 2018, 2019; Feng et al. 2018).
6.1 Limitations and future research
Various limitations of this study result from the available survey data. Most importantly, a sample of
225 investors is comparatively small, despite being based on a nationally representative data set of
over 3,800 persons. Consequently, the results may be unduly distorted by individual outliers. They
should therefore be interpreted with care and require substantiation by follow-up research. All surveys
suffer from the limitation that not everyone tells the truth. For example, very successful investors may
not want to disclose their returns in order to protect their identity or, at the opposite end of the
spectrum, ‘losers’ may hesitate to come out as such – even though the responses are anonymous.
Actual stock exchange data could yield more accurate results but do not appear to be available at
present.
The survey asks about the total amount invested in cryptocurrencies on the one hand and about the
year of the first investment in cryptocurrency on the other hand. This suggests that the results
presented in Table 1 should be critically scrutinized, since it is assumed that all investments were
made at the time of the initial investment in order to calculate performance against the market. Future
research should check the results based on a different data basis. The use of Bitcoin returns may be
subject to survivorship bias.
The questionnaire only lists the 15 largest currencies, yet four respondents additionally owned
Dogecoin. It is conceivable that the ownership ratio would be higher if Dogecoin had also been
included in the list due to a selection bias.
Our distinction between ‘short-term speculators’ and ‘long-term investors’ according to the
respondents’ investment horizon (see Table 1) provides only limited basis for further analysis. Future
research should therefore investigate long-horizon versus short-horizon investment motivations and
outcomes. The literature on individual stock investors provides a suitable basis for such an endeavor
12(Barber and Odean 2000, 2013). Corbet, Meegan, et al. (2018) find that investors with a short
investment horizon may gain diversification benefits by investing in multiple cryptocurrencies.
The literature on individual stock investors examines their trading behavior on the basis of the
amount and profitability of individuals’ trades. It examines what types of trades (limit vs. market
orders) investors use and to what extent fees influence individual performance. Kelley and Tetlock
(2013) find positive short-term returns for both order types, while Linnainmaa (2010) identifies
negative performance for limit orders and positive performance for market orders. These results differ
in Barber et al. (2009), where only limit orders yield positive short-term returns. Since cryptocurrency
markets resemble traditional stock markets in some respects but are quite different in others (e.g. 24/7
trading; decentralized trading venues etc.), it would be interesting to replicate this type of analysis for
cryptocurrency markets.
Cryptocurrency markets are extensively discussed in online and social media. There is a
considerable amount of disinformation, since for example various news outlets sell articles or
influencers on social media channels like Twitter advertise specific cryptocurrencies. The low level
of regulation means that there are much larger grey areas to exploit than in traditional markets.
Promising research topic include the analysis of media influence or disinformation, social media
sentiment and returns (Giannini et al. 2018), and social trading (Dorfleitner et al. 2018; Jin et al.
2019). The topic of social recognition of investment decisions that may result in increased individual
trading (Mensmann and Breitmayer 2018) can be tested for cryptocurrency markets.
7 Conclusion
This paper has analyzed the financial performance and socio-demographic characteristics of
cryptocurrency investors on the basis of an online survey that is representative of the German adult
internet-using population. The 225 individual investors we examined have generated an average
return of 300% over periods between one and ten years. However, similarly to other financial markets,
fewer than half of them (44%) were able to outperform the market. Based on their investment horizon,
we categorized 44% of the respondents as long-term investors and 27% as short-term investors. 77%
of the cryptocurrency investors are male, and they are 38 years old on average.
While we find that self-assessed knowledge about cryptocurrencies tends to promote larger returns
from cryptocurrency investments, the level of formal education has no significant effect. We therefore
suspect that the former effect is in fact attributable to reverse causation: Cryptocurrency investment
success breeds confidence in one’s own abilities, leading to greater self-ascribed knowledge in the
field, rather than other way around. Another interesting finding relates to the market dominance of
Bitcoin, which is owned by 82% of the investors, well ahead of Ethereum (29%). This accords with
our finding that the average investor owns no more than two different cryptocurrencies.
Our results provide a first scientific consideration of individual investors in cryptocurrencies and
may be used as a basis for a variety of future investigations. Investors seem to be aware of the risk
associated with cryptocurrencies, as evidenced by the fact that they invested only 60% of their
monthly net income – a survivable amount, even in the event of a total loss. In this respect, our study
may contribute to the ongoing debate about the risk and regulation of cryptocurrencies.
13References
Al Shehhi, A., Oudah, M., & Aung, Z. (2014). Investigating Factors behind Choosing a Cryptocurrency. In IEEE
international conference on industrial engineering and engineering management.
Anderson, A. (2013). Trading and under-diversification. Review of Finance, 17(5), 1699–1741. doi:10.1093/rof/rfs044
Ante, L. (2018). Cryptocurrency, Blockchain and Crime. In K. McCarthy (Ed.), The Money Laundering Market:
Regulating the Criminal Economy (pp. 171–198). Agenda Publishing. doi:10.2307/j.ctv5cg8z1.10
Ante, L. (2019). Market Reaction to Exchange Listings of Cryptocurrencies. doi:10.13140/RG.2.2.19924.76161
Ante, L., & Fiedler, I. (2019). Cheap Signals in Security Token Offerings (STOs). doi:10.2139/ssrn.3356303
Ante, L., Sandner, P., & Fiedler, I. (2018). Blockchain-Based ICOs: Pure Hype or the Dawn of a New Era of Startup
Financing? Journal of Risk and Financial Management, 11(4), 80. doi:10.3390/jrfm11040080
Barber, B. M., Lee, Y. T., Liu, Y. J., & Odean, T. (2009). Just how much do individual investors lose by trading?
Review of Financial Studies, 22(2), 609–632. doi:10.1093/rfs/hhn046
Barber, B. M., & Odean, T. (2000). Trading is hazardous to your wealth: The common stock investment performance of
individual investors. Journal of Finance, 55(2), 773–806. doi:10.1111/0022-1082.00226
Barber, B. M., & Odean, T. (2001). Boys will be boys: Gender, overconfidence, and common stock investment.
Quarterly Journal of Economics, 116(1), 261–292. doi:10.1162/003355301556400
Barber, B. M., & Odean, T. (2002). Online Investors: Do the Slow Die First? Review of Financial Studies, 15(2), 455–
487. doi:10.1093/rfs/15.2.455
Barber, B. M., & Odean, T. (2013). The Behavior of Individual Investors. In Handbook of the Economics of Finance
(Vol. 2, pp. 1533–1570). Elsevier B.V. doi:10.1016/B978-0-44-459406-8.00022-6
Baur, D. G., Hong, K. H., & Lee, A. D. (2018). Bitcoin: Medium of exchange or speculative assets? Journal of
International Financial Markets, Institutions and Money, 54, 177–189. doi:10.1016/j.intfin.2017.12.004
Blockchain Research Lab. (2019a). Perceived and Actual Use of Cryptocurrencies Results of a representative survey
among German Internet users. https://www.blockchainresearchlab.org/wp-content/uploads/2019/07/Report-2-
Perceived-and-actual-use-of-cryptocurrencies-1.pdf
Blockchain Research Lab. (2019b). The State of Cryptocurrency Adoption : Results of a representative survey among
German Internet users. https://www.blockchainresearchlab.org/wp-
content/uploads/2019/07/The_State_of_Cryptocurrency_Adoption_Q1_2019_Germany_Blockchain_Research_L
ab.pdf
Cheah, E. T., & Fry, J. (2015). Speculative bubbles in Bitcoin markets? An empirical investigation into the fundamental
value of Bitcoin. Economics Letters, 130, 32–36. doi:10.1016/j.econlet.2015.02.029
Corbet, S., Lucey, B., & Yarovaya, L. (2018). Datestamping the Bitcoin and Ethereum bubbles. Finance Research
Letters, 26, 81–88. doi:10.1016/j.frl.2017.12.006
Corbet, S., Meegan, A., Larkin, C., Lucey, B., & Yarovaya, L. (2018). Exploring the dynamic relationships between
cryptocurrencies and other financial assets. Economics Letters, 165, 28–34. doi:10.1016/j.econlet.2018.01.004
Dorfleitner, G., Fischer, L., Lung, C., Willmertinger, P., Stang, N., & Dietrich, N. (2018). To follow or not to follow –
An empirical analysis of the returns of actors on social trading platforms. Quarterly Review of Economics and
Finance, 70, 160–171. doi:10.1016/j.qref.2018.04.009
Døskeland, T. M., & Hvide, H. K. (2011). Do Individual Investors Have Asymmetric Information Based on Work
Experience? Journal of Finance, 66(3), 1011–1041. doi:10.1111/j.1540-6261.2011.01658.x
Dyhrberg, A. H. (2016). Bitcoin, gold and the dollar – A GARCH volatility analysis. Finance Research Letters, 16, 85–
92. doi:10.1016/j.frl.2015.10.008
Dyhrberg, A. H., Foley, S., & Svec, J. (2018). How investible is Bitcoin? Analyzing the liquidity and transaction costs
of Bitcoin markets. Economics Letters, 171, 140–143. doi:10.1016/j.econlet.2018.07.032
Fahlenbrach, R., & Frattaroli, M. (2019). ICO Investors. doi:10.2139/ssrn.3419944
Feng, W., Wang, Y., & Zhang, Z. (2018). Informed trading in the Bitcoin market. Finance Research Letters, 26, 63–70.
doi:10.1016/j.frl.2017.11.009
Fisch, C., Masiak, C., Vismara, S., & Block, J. (2019). Motives and profiles of ICO investors. Journal of Business
Research. doi:10.1016/j.jbusres.2019.07.036
French, K., & Poterba, J. (1991). Investor Diversification and International Equity Markets. The American Economic
Review. doi:10.2307/2006858
Giannini, R., Irvine, P., & Shu, T. (2018). Nonlocal Disadvantage: An Examination of Social Media Sentiment. The
Review of Asset Pricing Studies, 8(2), 293–336. doi:10.1093/rapstu/rax020
Gibbs, T., & Yordchim, S. (2014). Thai Perception on Litecoin Value. International Journal of Social, Education,
14Economics and Management Engineering, 8(8), 2589–2591.
Glaser, F., Zimmermann, K., Haferkorn, M., Weber, M. C., & Siering, M. (2014). Bitcoin - asset or currency?
revealing users’ hidden intentions. https://ssrn.com/abstract=2425247
Glaser, M., & Weber, M. (2007). Overconfidence and trading volume. Geneva Risk and Insurance Review, 32(1), 1–36.
doi:10.1007/s10713-007-0003-3
Goetzmann, W. N., & Kumar, A. (2008). Equity portfolio diversification. Review of Finance, 12(3), 433–463.
doi:10.1093/rof/rfn005
Graham, J. R., Harvey, C. R., & Huang, H. (2009). Investor competence, trading frequency, and home bias.
Management Science, 55(7), 1094–1106. doi:10.1287/mnsc.1090.1009
Grinblatt, M., & Keloharju, M. (2009). Sensation seeking, overconfidence, and trading activity. Journal of Finance,
64(2), 549–578. doi:10.1111/j.1540-6261.2009.01443.x
Grinblatt, M., Keloharju, M., & Linnainmaa, J. T. (2012). IQ, trading behavior, and performance. Journal of Financial
Economics, 104(2), 339–362. doi:10.1016/j.jfineco.2011.05.016
Hur, Y., Jeon, S., & Yoo, B. (2015). Is Bitcoin a Viable E-Business?: Empirical Analysis of the Digital Currency’s
Speculative Nature.
Ivković, Z., Sialm, C., & Weisbenner, S. (2008). Portfolio concentration and the performance of individual investors.
Journal of Financial and Quantitative Analysis, 43(3), 613–655. doi:10.1017/S0022109000004233
Jin, X., Zhu, Y., & Huang, Y. S. (2019). Losing by learning? A study of social trading platform. Finance Research
Letters, 28(April 2018), 171–179. doi:10.1016/j.frl.2018.04.017
Kahneman, D., & Tversky, A. (1979). Prospect Theory: An Analysis of Decision under Risk. Econometrica, 47(2),
263–292. doi:10.2307/1914185
Katsiampa, P. (2017). Volatility estimation for Bitcoin: A comparison of GARCH models. Economics Letters, 158, 3–6.
doi:10.1016/j.econlet.2017.06.023
Katsiampa, P. (2019). Volatility co-movement between Bitcoin and Ether. Finance Research Letters, 30, 221–227.
doi:10.1016/j.frl.2018.10.005
Kelley, E. K., & Tetlock, P. C. (2013). How Wise Are Crowds? Insights from Retail Orders and Stock Returns. Journal
of Finance, 68(3), 1229–1265. doi:10.1111/jofi.12028
Korniotis, G. M., & Kumar, A. (2011). Do older investors make better investment decisions? Review of Economics and
Statistics, 93(1), 244–265. doi:10.1162/REST_a_00053
Korniotis, G. M., & Kumar, A. (2013). Do portfolio distortions reflect superior information or psychological biases?
Journal of Financial and Quantitative Analysis, 48(1), 1–45. doi:10.1017/S0022109012000610
Kristoufek, L. (2018). On Bitcoin markets (in)efficiency and its evolution. Physica A: Statistical Mechanics and its
Applications, 503, 257–262. doi:10.1016/j.physa.2018.02.161
Linnainmaa, J. T. (2010). Do limit orders alter inferences about investor performance and behavior? Journal of
Finance, 65(4), 1473–1506. doi:10.1111/j.1540-6261.2010.01576.x
Mensmann, B., & Breitmayer, B. (2018). Social recognition and investor overcondence. doi:10.2139/ssrn.3140827
Mills, D. J., & Nower, L. (2019). Preliminary findings on cryptocurrency trading among regular gamblers: A new risk
for problem gambling? Addictive Behaviors, 92, 136–140. doi:10.1016/j.addbeh.2019.01.005
Nakamoto, S. (2008). Bitcoin: A Peer-to Peer Electronic Cash System. https://bitcoin.org/bitcoin.pdf. Accessed 12 June
2019
Powell, M., & Ansic, D. (1997). Gender differences in risk behaviour in financial decision-making: An experimental
analysis. Journal of Economic Psychology, 605–628. doi:10.1016/S0167-4870(97)00026-3
Schuh, S., & Shy, O. (2016). US consumers’ adoption and use of Bitcoin and other virtual currencies. In
DeNederlandsche bank: Retail payments: mapping out the road ahead.
Yermack, D. (2015). Is Bitcoin a Real Currency? Handbook of digital currency. doi:10.2139/ssrn.2361599
15Appendix
Table A1. Sample statistics and differences across sub-samples for demographics
I. II. III. IV.
Δ
Full No Positive Negative III. vs. IV.
sample change returns returns
Age groups
18-24 years 0.17 0.12 0.17 0.20 0.03
25-34 years 0.29 0.35 0.31 0.23 0.08
35-49 years 0.34 0.35 0.32 0.37 0.05
50+ years 0.20 0.18 0.20 0.20 0.00
Income groups
Under €500 0.02 0.03 0.02 0.02 0.00
€500 to €999 0.04 0.09 0.02 0.05 0.03
€1,000 to €1,499 0.13 0.09 0.09 0.25 0.16***
€1,500 to €1,999 0.17 0.21 0.15 0.18 0.03
€2,000 to €2,999 0.22 0.21 0.26 0.15 0.11*
€3,000 to €4,999 0.32 0.32 0.36 0.26 0.10
Over €5,000 0.08 0.06 0.10 0.08 0.02
Education levels
No school leaving certificate 0.01 0.03 0.00 0.02 0.02
Lower secondary school 0.09 0.12 0.10 0.06 0.04
High school 0.20 0.18 0.20 0.20 0.00
Commercial training 0.12 0.21 0.11 0.09 0.01
Trade training 0.21 0.18 0.19 0.26 0.07
University degree 0.34 0.26 0.38 0.35 0.03
PhD 0.04 0.03 0.05 0.02 0.03
Observations 225 34 126 65
***, * indicates significance at the 1% and 10% level, respectively (t-test).
16Table A2. Descriptive statistics, correlations and variance inflation factors
Variables N Mean SD p50 Min Max (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (17) (18) VIF
Dependent variables
(1) Return (log) 225 1.23 1.05 0.85 0.01 6.50 1.00 1.81
(2) Positive 225 0.56 - 1 0 1 0.63 1.00 2.32
(3) Negative 225 0.29 - 0 0 1 -0.52 -0.72 1.00 2.30
Industry knowledge
(4) Crypto knowledge 225 7.58 1.79 8 1 10 0.16 0.16 -0.12 1.00 1.73
(5) Coins known 225 5.92 3.48 5 1 15 -0.01 0.08 0.01 0.22 1.00 1.40
(6) Exchanges 225 1.88 2.14 1 0 19 0.07 0.13 -0.04 0.23 0.28 1.00 1.83
Attitudes
(7) Trust 225 6.79 2.11 7 0 10 0.15 0.12 -0.16 0.50 0.12 0.25 1.00 1.45
(8) Ideology 225 6.16 2.65 7 0 10 0.10 0.13 -0.18 0.32 0.03 0.31 0.46 1.00 2.14
Demographics
(9) Age 225 37.64 13.80 35 18 85 0.03 -0.03 0.01 -0.07 -0.13 -0.15 -0.22 -0.21 1.00 1.17
(10) Male 225 0.77 - 1 0 1 -0.02 -0.01 0.06 0.16 0.08 -0.11 -0.02 0.00 0.03 1.00 1.16
(11) Income 223 2.93 1.56 2.50 0.45 6.5 0.12 0.14 -0.12 0.27 0.17 0.22 0.15 0.13 0.02 0.09 1.00 1.27
(12) Education 225 3.01 1.06 3 0 5 0.03 0.05 0.00 0.20 0.12 0.01 0.00 -0.01 0.05 0.19 0.34 1.00 1.27
Diversification
(18) Coins owned 225 2.02 1.67 1 0 15 0.07 0.04 0.02 0.19 0.45 0.58 0.15 0.15 -0.18 -0.06 0.13 0.07 1.00 2.16
(19) Coins / invest (log) 221 -5.39 1.96 -5.52 -10.82 0.41 0.17 0.01 -0.08 -0.14 -0.05 0.02 -0.02 -0.09 -0.04 -0.24 -0.29 -0.23 0.10 1.00 1.28
Correlations that are significant at the 5%-level are highlighted in bold.
17Declarations
Availability of data and materials
The datasets used and/or analyzed during the current study are
available from the corresponding author on request.
Conflicts of interest
Not applicable.
Funding
Not applicable.
Acknowledgements
Not applicable.
About the Blockchain Research Lab
The Blockchain Research Lab promotes independent science and
research on blockchain technologies and the publication of the
results in the form of scientific papers and contributions to
conferences and other media. The BRL is a non-profit
organization aiming, on the one hand, to further the general
understanding of the blockchain technology and, on the other
hand, to analyze the resulting challenges and opportunities as well
as their socio-economic consequences.
www.blockchainresearchlab.org
18You can also read
